← Cockpit
CMQ_042predictionAI/ComputeCPU-bottleneck-shift

As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory.

Predictor: Morgan Stanley

Prior probability
72.0%
Current probability
34.9%
evolves via intake + LBP
Conviction
5/5
Signal quality
A
Resolution
in_progress
Window
2026-01-01 – 2026-12-31
Edges in / out
1 / 0
Tickers exposed
26

Prediction text

As AI evolves from Generative to Agentic, structural computing bottlenecks shift away from GPU and heavily toward CPU and system memory. | Server CPU TAM + agentic workload profiling

Key catalyst: Server CPU TAM + agentic workload profiling

Watch events: CPU utilization % in agentic workflows; server CPU TAM re-rating by sell-side.

Resolution evidence

Status: in_progress

Agentic workflows in production (Anthropic Claude agents, Cognition Devin, OpenAI agents) show measurable CPU-bound latency patterns.

Predictor: Morgan Stanley

κ + Brier as of 2026-05-22
κ (discount)
0.633
Brier
0.0442
excellent
Hits / Misses
1 / 0
of 2 resolved
Hit rate
50.0%
Calibration plot (stated vs observed)

Evidence about this node from Morgan Stanley is multiplied by κ in /api/intake. Lower κ = less weight; floors at 0.10 (effectively silenced) and caps at 1.00 (full weight).

Reference class

Not linked

This node isn't linked to a reference class. The Bayesian update applies without outside-view blending.

Probability over time

6 prob_history rows
0%25%50%75%100%prior 72%2026-05-022026-05-172026-05-30
intake v2milestone miss sweeplbp propagationreference class assignedlegacy v1prior_prob (analyst seed)current = 34.9%

Milestone chain

Pre-event signals (upstream prereqs + window checkpoints) → resolution event → downstream cascades. Status/dates update from linked nodes; re-derive nightly via scripts/ops/derive_milestones.py.
Leading chain: 2 fired ✓ · 2 overdue ⏱ · 2 pending
  1. 2026-03-07overdueQ1 window check-in (25%)
  2. 2026-04-20hitMorgan Stanley publishes formal Rise of AI Agents report — GPU-to-CPU pivot thesis
    How: Morgan Stanley publishes major research report formally identifying agentic AI shift from GPUs to CPUs and memory as structural
    Source: Bitget News / Morgan Stanley — The Rise of AI Agents (April 2026)conf 99%
  3. 2026-05-12overdueQ2 window check-in (50%)
  4. 2026-04-22hitCPU-side orchestration is empirically measured at 50-90% of agentic workload latency
    How: Published benchmark or analyst report empirically measures CPU-side orchestration as 50-90% of agentic workload latency
    Source: ANI News — Agentic AI shifts value from GPUs to CPUs and memoryconf 90%
    Notes: Already cited by Morgan Stanley April 2026 — empirical CPU-bottleneck data live.
  5. 2026-07-17pendingQ3 window check-in (75%)
  6. 2026-04-01 → 2026-12-31pendingHyperscaler greenfield CPU racks deployed exclusively for agentic AI
    How: At least one hyperscaler (Meta, Google, Microsoft, AWS) publicly discloses standalone CPU rack deployment dedicated to agentic AI orchestration
    Source: BigGo Finance — Morgan Stanley: AI Agents Drive CPU Demandconf 85%
    Notes: NVIDIA standalone CPU at Meta already counts as initial signal.
  7. 2026-12-01 → 2027-12-31pendingMemory and CPU vendors outperform GPU pure-plays in 2026-27 returns
    How: Memory makers (Micron, SK Hynix, Samsung) + server CPU vendors (Intel, AMD, Arm) cumulative 12-mo total return exceeds NVIDIA over 2026-27 in any 12-mo window
    Source: Morgan Stanley — Why Doubling Down on Memory Stocks Amid AI Boomconf 50%
  8. 2027-01-01 → 2030-12-31pendingDRAM demand from agentic workloads adds 15-45 EB by 2030
    How: Morgan Stanley / IDC track agentic-driven incremental DRAM demand reaching 15-45 exabytes by 2030
    Source: Morgan Stanley — Agentic AI shifts value from GPUs to CPUs and memoryconf 70%
  9. 2028-01-01 → 2030-12-31pending$32.5-60B incremental CPU TAM materialized by 2030 per Morgan Stanley
    How: Server CPU TAM grows by ≥$30B incremental over 2026 baseline by 2030 attributable to agentic AI workloads
    Source: NewKerala — Agentic AI Creates $60B CPU Market by 2030: Morgan Stanleyconf 65%

What if this resolves?

Clamp this prediction TRUE or FALSE and run a counterfactual Gibbs sample. Surfaces the predictions whose marginals shift most under that assumption.
(live posterior: 35%)

Click a button to clamp this prediction and run a Gibbs sample. Returns the predictions whose marginals shift most. ~30s per run; ideal for stress-testing "if X resolves, what else moves?"

Evidence chain

Every probability update with full Bayesian provenance — chronological, latest first
metadata_milestone_miss_sweep2026-05-30T22:15:00Z34.9%-6.0pp
metadata_milestone_miss_sweep bayesian_v2 n=1 inside=0.349 blend=0.349 LLR=-0.257 κ=0.63 no_blend
Raw metadata
{
  "trf": 0.5881123843731557,
  "kappa": 0.6333,
  "base_rate": null,
  "predictor": "Morgan Stanley",
  "total_llr": -0.4054651081081644,
  "grace_days": 7,
  "bayesian_v2": true,
  "prior_logit": -0.3657020103529956,
  "bayes_factor": "1.3:1 against",
  "blend_reason": "no reference_class linked",
  "inside_prior": 0.409579974194961,
  "kappa_source": "predictor_table",
  "n_milestones": 1,
  "blend_applied": false,
  "contributions": [
    {
      "llr": -0.4054651081081644,
      "kind": "quartile_checkpoint",
      "kappa": 0.6333,
      "label": "Q2 window check-in (50%)",
      "weight": 0.05,
      "strength": "weak",
      "confidence": null,
      "source_url": null,
      "adjusted_llr": -0.2567810529649005,
      "expected_date": "2026-05-12",
      "measurement_criterion": null
    }
  ],
  "evidence_kind": "metadata_milestone_miss_sweep",
  "inside_source": "history_v2",
  "inside_weight": 0.5883213309387909,
  "outside_weight": 0.4116786690612091,
  "posterior_prob": 0.3492169282993375,
  "posterior_logit": -0.6224830633178962,
  "predictor_brier": 0.0442,
  "inside_posterior": 0.3492169282993375,
  "blended_posterior": 0.3492169282993375,
  "reference_class_id": null,
  "total_adjusted_llr": -0.2567810529649005,
  "predictor_n_resolved": 2
}
LBP2026-05-24T02:00:02Z41.0%-1.8pp
Network propagation: 42.7% → 41.0%
4-iter LBP, residual 0.01000 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run 806b02f8
LBP2026-05-17T02:00:01Z42.7%-3.6pp
Network propagation: 46.3% → 42.7%
5-iter LBP, residual 0.00689 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run e607fa96
LBP2026-05-10T02:00:02Z46.3%-7.2pp
Network propagation: 53.5% → 46.3%
6-iter LBP, residual 0.00584 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run e5c18d29
LBP2026-05-03T02:00:01Z53.5%-13.5pp
Network propagation: 67.0% → 53.5%
6-iter LBP, residual 0.00677 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run 1a683ac9
metadata_milestone_miss_sweep2026-05-02T22:07:21Z67.0%-5.0pp
metadata_milestone_miss_sweep bayesian_v2 n=1 inside=0.670 blend=0.670 LLR=-0.237 κ=0.58 no_blend
Raw metadata
{
  "trf": 0.6650500679109432,
  "kappa": 0.5833,
  "base_rate": null,
  "predictor": "Morgan Stanley",
  "total_llr": -0.4054651081081644,
  "grace_days": 7,
  "bayesian_v2": true,
  "prior_logit": 0.9444616088408513,
  "bayes_factor": "1.3:1 against",
  "blend_reason": "no reference_class linked",
  "inside_prior": 0.72,
  "kappa_source": "predictor_table",
  "n_milestones": 1,
  "blend_applied": false,
  "contributions": [
    {
      "llr": -0.4054651081081644,
      "kind": "quartile_checkpoint",
      "kappa": 0.5833,
      "label": "Q1 window check-in (25%)",
      "weight": 0.05,
      "strength": "weak",
      "confidence": null,
      "source_url": null,
      "adjusted_llr": -0.2365077975594923,
      "expected_date": "2026-03-07",
      "measurement_criterion": null
    }
  ],
  "evidence_kind": "metadata_milestone_miss_sweep",
  "inside_source": "prior_prob",
  "inside_weight": 0.5344649524623397,
  "outside_weight": 0.4655350475376603,
  "posterior_prob": 0.6699488693573884,
  "posterior_logit": 0.707953811281359,
  "predictor_brier": 0.01,
  "inside_posterior": 0.6699488693573884,
  "blended_posterior": 0.6699488693573884,
  "reference_class_id": null,
  "total_adjusted_llr": -0.2365077975594923,
  "predictor_n_resolved": 1
}

Network propagation neighbors

Top edges sorted by latest LBP cross-impact
All propagation →

No propagation data yet. Run inference/.venv/bin/python scripts/ops/run_loopy_belief_propagation.py on the droplet, or wait for the Sunday 02:00 UTC weekly cron.

Ticker exposure

26 ticker(s) linked

Beneficiaries (22)

TSMAMBAARMCEVACRWVDOCNIRENNBISNVDASITMALABCSCODELLSIEGYINTCNXPILNVGYAMDANETMRVLAVGOQCOM

Prerequisites (1)

Predictions that must hit first
TypePredTitleDomainLag
correlateS_COMPUTE_100GW_2030Compute: 100GW national-scale by Dec 2030compute_scale

Dependents (0)

Predictions enabled by this
TypePredTitleDomainLag
No dependents

Validations (1)

Resolution events
Observed atStatusByNotes
2026-04-29partialthesis_timeline_v1.0_importAgentic workflows in production (Anthropic Claude agents, Cognition Devin, OpenAI agents) show measurable CPU-bound latency patterns.

Linked documents (10)

Auto-generated by cosine similarity from Polymarket / Manifold / EDGAR / GDELT

Raw metadata

From Thesis_Timeline_v1.0_FINAL workbook
{
  "nia": false,
  "qty": "GPU→CPU shift",
  "mode": "THESIS",
  "role": "Cited-Firm",
  "context": "Core thesis of MS 73-page 'Rise of the AI Agent – Global Implications' report; contrarian vs consensus GPU-centric view.",
  "to_year": 2030,
  "cited_by": "Rise of AI Agent report",
  "conv_cues": "conclude; heavy thesis weight",
  "direction": "HAPPEN",
  "from_year": 2026,
  "timeframe": "2026+",
  "conv_level": "HIGH",
  "milestones": [
    {
      "kind": "quartile_checkpoint",
      "label": "Q1 window check-in (25%)",
      "status": "overdue",
      "weight": 0.05,
      "ordinal": -6,
      "source_id": null,
      "expected_date": "2026-03-07",
      "observed_date": null,
      "miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
      "miss_emitted_by": "metadata_milestone_sweep"
    },
    {
      "kind": "llm_pre_event",
      "label": "Morgan Stanley publishes formal Rise of AI Agents report — GPU-to-CPU pivot thesis",
      "source": "Bitget News / Morgan Stanley — The Rise of AI Agents (April 2026)",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -5,
      "source_id": null,
      "confidence": 0.99,
      "source_url": "https://www.bitget.com/amp/news/detail/12560605375963",
      "expected_date": "2026-04-20",
      "observed_date": "2026-04-20",
      "research_origin": "deep_research",
      "measurement_criterion": "Morgan Stanley publishes major research report formally identifying agentic AI shift from GPUs to CPUs and memory as structural"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q2 window check-in (50%)",
      "status": "overdue",
      "weight": 0.05,
      "ordinal": -4,
      "source_id": null,
      "expected_date": "2026-05-12",
      "observed_date": null,
      "miss_emitted_at": "2026-05-30T22:15:00.756418+00:00",
      "miss_emitted_by": "metadata_milestone_sweep"
    },
    {
      "kind": "llm_pre_event",
      "label": "CPU-side orchestration is empirically measured at 50-90% of agentic workload latency",
      "notes": "Already cited by Morgan Stanley April 2026 — empirical CPU-bottleneck data live.",
      "source": "ANI News — Agentic AI shifts value from GPUs to CPUs and memory",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -3,
      "source_id": null,
      "confidence": 0.9,
      "source_url": "https://aninews.in/news/business/morgan-stanley-agentic-ai-shifts-value-from-gpus-to-cpus-and-memory-creating-up-to-60bn-incremental-cpu-tam-by-203020260422131744/",
      "expected_date": "2026-07-01",
      "observed_date": "2026-04-22",
      "research_origin": "deep_research",
      "expected_date_range": {
        "to": "2026-09-30",
        "from": "2026-04-01"
      },
      "measurement_criterion": "Published benchmark or analyst report empirically measures CPU-side orchestration as 50-90% of agentic workload latency"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q3 window check-in (75%)",
      "status": "pending",
      "weight": 0.05,
      "ordinal": -2,
      "source_id": null,
      "expected_date": "2026-07-17",
      "observed_date": null
    },
    {
      "kind": "llm_pre_event",
      "label": "Hyperscaler greenfield CPU racks deployed exclusively for agentic AI",
      "notes": "NVIDIA standalone CPU at Meta already counts as initial signal.",
      "source": "BigGo Finance — Morgan Stanley: AI Agents Drive CPU Demand",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -1,
      "source_id": null,
      "confidence": 0.85,
      "source_url": "https://finance.biggo.com/news/-ffAq50BvthpMgHB4o2T",
      "expected_date": "2026-08-16",
      "research_origin": "deep_research",
      "expected_date_range": {
        "to": "2026-12-31",
        "from": "2026-04-01"
      },
      "measurement_criterion": "At least one hyperscaler (Meta, Google, Microsoft, AWS) publicly discloses standalone CPU rack deployment dedicated to agentic AI orchestration"
    },
    {
      "kind": "event",
   
... (truncated)